yielding to high yields? critiquing food security...

Geoforum 71 (2016) 33–43

Contents lists available at ScienceDirect

Geoforum

journal homepage: www.elsevier .com/locate /geoforum

Yielding to high yields? Critiquing food security definitions and policyimplications for ethnic minority livelihoods in upland Vietnam

http://dx.doi.org/10.1016/j.geoforum.2016.03.0010016-7185/� 2016 Elsevier Ltd. All rights reserved.

⇑ Corresponding author.E-mail addresses: [email protected] (V. Kyeyune), sarah.turner@

mcgill.ca (S. Turner).

1 The 2010 poverty line was VND 3.12 million per capita per year (US$162Clearly, poverty rates are an incomplete story, given that many uplanminorities maintain semi-subsistence livelihoods.

Victoria Kyeyune ⇑, Sarah TurnerDepartment of Geography, McGill University, 805 Sherbrooke Street West, Montreal, Quebec H3A 0B9, Canada

a r t i c l e i n f o

Article history:Received 3 August 2015Received in revised form 20 February 2016Accepted 2 March 2016Available online 15 March 2016

Keywords:Hybrid seedsMaizeFood securityLivelihoodsEthnic minoritiesVietnam

a b s t r a c t

Agricultural intensification is at the core of the current agrarian transition in Southeast Asia. New cropvarieties promise higher productive outputs, but depend on significant increases in chemical inputs. Inthe highlands of northern Vietnam, we find that adopting hybrid maize is inevitably associated withan increasing dependence on cash for direct and indirect inputs and investments. This reliance on thecash economy is a new reality for semi-subsistence ethnic minority Hmong households, and provides evi-dence of the advancing agrarian transition in Vietnam’s remote northern highlands. While livelihood out-comes of hybrid seed adoption include increased maize yields, local farmers highlight numerousdrawbacks, including unstable input prices, limited storage periods, pest concerns, and the increased reli-ance on cash. Strong preferences for the taste of traditional local maize, as well as concerns over regularharvests, lead many households to resist the full adoption of new hybrid varieties and redirect hybridmaize to livestock feed and household alcohol production instead. Thus while state policies extoll the vir-tues of high-yielding hybrid maize for poverty reduction, we find that food availability is an overempha-sized element of household food security and upland agricultural development policies. Food securityinterventions must move beyond conceptualizing food security as a result of food availability alone,and also incorporate cultural acceptability of food, better understandings of hybrid maize cultivationchallenges, and respect the seed diversity on which local livelihoods and food security rely.

� 2016 Elsevier Ltd. All rights reserved.

1. Introduction food security, biodiversity, environmental degradation, and cli-

Noteworthy transformations are taking place throughout ruralSoutheast Asia linked to increasing market integration, the unre-lenting commoditization of production, land grabbing, and the clo-sure of land frontiers (Nevins and Peluso, 2008; Borras et al., 2011;Hall et al., 2013). With increasing demand for timber, biofuels, andfood products, substantial land conversions are taking place along-side an unprecedented intensification of rural land use. Numerousactors from individual smallholders to large transnational corpora-tions are involved, with these dynamics combining to intensify anddeepen the agrarian transition at a scale unprecedented in South-east Asia. Some individuals and households have benefitted signif-icantly from increased commoditization and global market links,others have engaged only selectively in new market opportunities,while further groups have been disadvantaged by infringementsupon indigenous rights, reduced resource access, and intensifyingcultural clashes (Moore, 1998; Caouette and Turner, 2009). Thesesweeping changes are also closely connected to the challenges of

mate change (Misselhorn et al., 2012).In the midst of this agrarian transition, Vietnam has been con-

sidered a poverty reduction and development success story afterreaching lower middle-income country status in 2009. Surpassinga number of poverty-related Millennium Development Goals, thenational poverty rate has fallen from 58% in 1993 to less than 3%in 2015 (World Bank, 2009a, 2012, 2015). Moreover, Vietnamnow fluctuates between being the world’s third or fourth largestrice exporter. While this all sounds extremely positive at the macrolevel, there are still real and increasing disparities in socio-economic wellbeing and persistent food security challenges(Nguyen Viet Cuong et al., 2015). Notably, ethnic minority house-holds in Vietnam’s northern uplands are seldom reaping therewards of the country’s agrarian transition and economic growth.The 2010 Vietnam Household Living Standard Survey (VHLSS) indi-cated that poverty rates were highest in the Northern Midland andMountains region (29.4%), with ethnic minority Hmong house-holds among the poorest at 48.7% (GSO, 2010a).1 While the govern-

in 2010).d ethnic

34 V. Kyeyune, S. Turner / Geoforum 71 (2016) 33–43

ment has implemented upland poverty reduction and food securitypolicies, little is known about the impacts of these policies on localpeople’s long-held livelihood strategies and approaches to foodsecurity.

In its quest to enhance national food security, Vietnam’s gov-ernment is promoting intensified maize cultivation. Nationalmaize output rose significantly between 1961 and 2004(Gerpacio and Pingali, 2007), and this trend continued with 2004production levels at 3.43 million tonnes (yield: 34, 617 ha) risingto over 5.19 million tonnes (yield: 44,354 ha) in 2013 (FAOSTAT,2015). This rise has been buoyed by technological innovations suchas high-yielding hybrid seeds,2 strongly endorsed by the govern-ment. Production is further encouraged to feed the domestic ‘live-stock revolution’ (Gerpacio and Pingali, 2007; FAOSTAT, 2015).Concurrently, maize remains a staple crop for many upland subsis-tence producers and, due in part to subsidies and state propaganda,3

hybrid maize varieties have been widely adopted to replace loweryielding locally-bred or ‘traditional’ varieties (landraces), includingwild, open pollinated varieties (Zeven, 1998; Dang Thanh Ha et al.,2004; Erenstein, 2010; Stromberg et al., 2010). This continuing risein demand for maize for food and animal feed is expected to resultin further crop substitution, commercialization of existing maize-based production systems, and expanded and intensified maize cul-tivation, particularly in the agriculturally marginalized northernuplands (Gerpacio and Pingali, 2007).

Approximately 6.2 million people in small-farm households inVietnam’s northern region are food insecure or at risk of food inse-curity (FAO, 2004). Hmong ethnic minority households in the fourmost mountainous border provinces (Sơn La, Lai Châu, Hà Giang,and Cao Bằng) are considered among the poorest and most vulner-able to food insecurity (FAO, 2004; Minot et al., 2006). Given thepace of the agrarian transition occurring throughout the region,researchers have suggested that the rapid modernization and com-mercialization of upland farming in such remote areas is unlikely toreduce poverty for many reasons – including inadequate marketinstitutions, insufficient insurance, and weak communal safetynets (Pandey et al., 2006). Through our own fieldwork we havefound that Hmong livelihoods in Hà Giang province rely heavilyon maize cropping systems. However, limited scholarship isdevoted to food security and livelihoods in this province, and todate, none focuses on the challenges and constraints to localHmong household food security as Vietnam advances its agendafor industrialized agriculture.

This paper investigates how ethnic minority Hmong householdsin Hà Giang province, northern Vietnam (Fig. 1), are adapting to theintroduction of state-supported hybrid maize seeds and theimpacts this agricultural intensification program is having on localfood security. We have three core objectives. First, to understandthe role of state actors in the introduction and adoption of hybridmaize seeds in Hà Giang province. Second, to investigate the suc-cesses and challenges for Hmong household livelihoods adoptinghybrid maize. Third, to examine the degree of agency Hmong farm-ers reveal in their negotiations over hybrid maize adoption. But

2 Hybrid seeds are produced from selectively crossing two inbred and geneticallydifferent parental (P) lines. Female plants of the first (F1) offspring generationproduce hybrid seeds, which are planted. The resulting hybrid plants possessheterosis or ‘hybrid vigor’, tending to produce higher yields. Yet, the seeds from thesehybrid plants are significantly less productive if planted again; therefore farmersmust purchase new hybrid seeds yearly (IRRI 2015). By contrast, random open-pollinated varieties (OPVs), and farmer-developed crop varieties (farmer-saved seeds)produce biodiverse landraces adapted to local environments (Zeven 1998; Gerpacio2001; Vernooy 2003). A central problem of hybrid seeds in the paradigm ofsustainable agriculture is the fact that they are essentially a genetic ‘dead-end’.

3 State propaganda to promote hybrid maize seed adoption includes postersextolling the virtues of hybrid maize pasted on the sides of road-side stalls and statebuildings throughout the uplands, announcements made over village loudspeakersystems, and calendars distributed to farmers.

before we address these objectives, why is this not just anotherpaper on the Green Revolution? First, hybrid varieties (see footnote2) are not the same as the original Green Revolution high yieldvarieties (HYVs) of wheat and rice. Second, while some uplandsemi-subsistence farmers in this case study sell hybrid maize forlivestock feed or as maize alcohol, rather than consuming it, thismarket involvement is nowhere near the same degree as farmerstypically targeted by Green Revolution technologies (hence a dis-tinction from lowland Vietnam and many other areas of Asia).Third, as Bonnin and Turner (2012) found for hybrid rice in a neigh-boring upland province, the introduction of hybrid seeds and asso-ciated technologies is not presently leading to important socialdifferentiation as occurred earlier in India with HYVs.

To address our objectives, we next propose a conceptual frame-work that draws on food security, sustainable livelihoods, andagency debates, before introducing the actors at the core of thisinvestigation. We examine how the hybrid maize program hasbeen implemented in these uplands, the official actors involved,and their understandings of the benefits and drawbacks for localcommunities. We then turn to local farmer interpretations of thisintervention, focusing on four causal factors of livelihood vulnera-bility identified during interviews. We find that farmers are flexingtheir agency in the face of state agricultural programs, determiningthe limits of adoption with regards to their astute knowledge ofenvironmental limitations as well as cultural acceptability. Accord-ingly, we argue that food security interventions must move beyondcommonly conceptualizing food security as a result of food avail-ability alone. Consideration must also be paid to the local seeddiversity on which livelihoods and food security rely, cultivationchallenges, and to more fully recognizing the importance of cultur-ally acceptable food.

This research isbasedondatagatheredduringfield research inHàGiang province. During May–July 2013 the first author completedunstructured/conversational interviews with 51 Hmong maize cul-tivators in six communes (Ðồng Van, Phố Cáo, Phó Bảng, Tả Phìn,Thài Phìn Tủng and Lung Cú), in Ðồng Van district, Hà Giang, regard-ing their adoption (or not) of hybrid seeds and associated livelihooddilemmas. Twelve semi-structured interviews were also completedwith agricultural extension officers, state officials, and NGO repre-sentatives. These data are supplemented by information gatheredin 2009 and 2010 by the second author during interviews with overthirtyHmong farmers andmarketplace traders inHàGiang provinceregarding rural livelihoods.

2. Conceptualizing food security, sustainable livelihoods, andthe role of agency

Understanding the core dimensions of food security is critical toexamining the impacts of state policies on semi-subsistence agri-cultural livelihoods in Vietnam’s northern uplands. Food securitycan be considered an ideology, reflecting our normative sensitivi-ties about hunger, inequalities in access to food and the means toproduce it, and power differentials in the food system (Maslow,1954; Maxwell, 1996). Maxwell (1996) has explored the develop-ment of post-modern undercurrents in food security debates dur-ing the decades following the 1974 World Food Conference when‘food security’ was defined as the ‘‘availability at all times of ade-quate world food supplies of basic foodstuffs to sustain a steadyexpansion of food consumption and to offset fluctuations in pro-duction and prices” (UN, 1975, 14). Since then, Maxwell (1996)has identified three main transitions: from global and national tohousehold and individual scales; from a ‘food first’ perspectiveviewing food as a lower-order human need to a preoccupation withlong-term livelihood security (resilience) as a necessary condition

Fig. 1. Hà Giang province situated in northern Vietnam with field site communes shaded (Map credit: Ðinh Thị Diệu).

V. Kyeyune, S. Turner / Geoforum 71 (2016) 33–43 35

for food security; and from objective indicators to subjective per-ceptions of causes and situations leading to food insecurity.

Today, numerous definitions of food security revolve arounddimensions including availability, acceptability, accessibility, appro-priateness, and agency (Pinstrup-Andersen, 2009; FAO, 2015;WHO, 2015). While many countries and NGOs continue to empha-size availability as their key food security focus, debates have arisenover how to create more inclusive, actor-oriented approaches tofood security (Jacobson, 2007). It has been argued that definingand analyzing food security at the individual level requires a moresubjective approach, while acknowledging that individual foodsecurity is still influenced by broader political and socioeconomiccontexts (Maxwell, 1996). Indeed, it is only by accentuating thevoices and lived experiences of individual actors and their under-standings of food security and sustainable livelihoods, that onecan truly begin to understand the local, everyday realities of creat-ing and maintaining food secure livelihoods that are culturallyacceptable and appropriate (Kontinen, 2004; Long, 2004). In thispaper we show that analyses of individual and household foodsecurity require greater attention to qualitative measures – suchas cultural acceptability of food, perceptions of well-being, andthe political processes that influence access to food. This ensuresthat food security interventions do not follow an oversimplifiedlinear trajectory, as cautioned by Long and Van Der Ploeg (1989).

To investigate the challenges for Hmong farmers adoptinghybrid maize, we also draw on the sustainable livelihoods litera-ture. Chambers and Conway (1992, 5) initially defined livelihoodsas being comprised of ‘‘capabilities, assets (including both materialand social resources) and activities required for a means of living”.They added that a sustainable livelihood can ‘‘cope with andrecover from stresses and shocks, maintain or enhance its capabil-ities and assets, while not undermining the natural resource base”(Chambers and Conway, 1992, 5). A sustainable livelihoodapproach emphasizes that livelihood assets (financial, natural,physical, social, human) are a means to achieving various liveli-hood outcomes, including improved food security, while notingthat livelihood strategies4 are vulnerable to external transformingstructures and processes, and social exclusion. The vulnerability con-

4 Scoones (1998, 9) defines three broad clusters of livelihood strategies: agricul-tural intensification or extensification, livelihood diversification, and migration.

text is especially acute for livelihoods depending on semi-subsistence agriculture. As we explore in our research, governmentpolicies towards ethnic minorities in Vietnam’s northern uplandsaim to transform local agricultural practices, thereby influencingthe livelihood vulnerability context of rural households. In thisregard, Scoones (2009) has also emphasized the need for more polit-ical analysis of livelihood perspectives, particularly relevant for ourresearch with ethnic minorities in socialist Vietnam, where accessto arable land and inputs for intensified agriculture (hybrid seeds,chemical fertilizers, and pesticides) is influenced by state policies.

By underplaying less tangible social and political influences inpractice, the sustainable livelihoods framework can be constrainedby its tendency to focus on aspects of material access and ability(Kanji et al., 2005; Scoones, 2009; Carr, 2013). Yet, ever since KarlPolanyi’s seminal works (e.g., Polanyi, 1944), many social anthro-pologists and human geographers have demanded the recognitionof ‘‘the cultural, historical, and spatial dynamics of rural liveli-hoods—in addition to the more obvious economic dynamics”(McSweeney, 2004, 638). These dynamics, embedded in local his-tories, customs, and systems of regulation, in turn shape economicexchanges and related livelihood decision-making (de Haan andZoomers, 2003, 2005; Eakin et al., 2006). By considering suchmicro-scale social relations and their embeddedness, more inclu-sive and culturally appropriate actor-oriented approaches to liveli-hoods are possible. Indeed, if we listen to the voices andexperiences of individual actors, respect homegrown knowledgesof ‘development’, and analyze the way local actors adapt moderncircumstances to their reality, we can shed light on the local every-day practicalities of how people use their agency to make, negoti-ate, and protect their livings (McGregor, 2009; Turner et al., 2015).

3. Situating people, politics, and agrarian change in uplandnorthern Vietnam

Vietnam is an ethnically diverse society. The state recognizes 54‘nationalities’; the Kinh majority (Việt, lowland Vietnamese), and53 ‘national minorities’ (các dân tộc thiểu số). Hmong are the sec-ond largest ethnic minority group in the northern uplands, num-bering just over 1 million. In Hà Giang province, the mostnorthern province in Vietnam, sharing a border with Yunnan,China, the total population is 724,537, of whom 231,464 are


Hmong, the province’s largest ethnic group (GSO, 2010b). Here,semi-subsistence Hmong households engage primarily in cultivat-ing staple food crops (maize and rice), various vegetables and fruitin small home gardens, and raising livestock. In the northern dis-trict of Ðồng Van, where this study takes place, tough agro-ecological conditions including a steep karst landscape and poorsoils, are suited to cultivating upland maize, rather than rice.

Before and after the introduction of Ðổi mới, the state’s eco-nomic renovation that began in the mid-1980s, numerous statestrategies were designed to ‘‘handle highland minorities in themost effective and economic way” to avoid impeding Vietnam’ssteady ‘modernization’ (Michaud, 2009, 25). By 1998, the Vietnamgovernment had established 21 different national policies and pro-jects aimed at socioeconomic development, cultural moderniza-tion, and poverty reduction among ‘backwards’ ethnic minoritiesin upland areas. Such state policies continue to impact the locallivelihood and agricultural choices available.

The most important poverty reduction program for residents inwhat are labeled ‘remote and difficult areas’ is Program 135 or‘Socio-economic Development for the Communes Facing GreatestHardships in the Ethnic Minority and Mountainous Areas’. Firstintroduced in 2001 and implemented in three stages to date, Pro-gram 135 was designed to ‘‘help people in ethnic minority andmountainous areas overcome poverty, narrow the income gap withother communes in other districts and provinces and eliminaterisks for social instability” (Ha Viet Quan, 2009, 2; see alsoNguyen Viet Cuong et al., 2015). A key component is the provisionof agricultural extension and technology, such as access toimproved varieties of seed (World Bank, 2009a, 2009b; Bonninand Turner, 2012). Nevertheless, upland ethnic minorities still fre-quently encounter government officials lacking reliable informa-tion about their livelihoods and cultures, with the state pushingto integrate them into the national agenda of neo-liberal growth(Forsyth and Michaud, 2011). This policy framework has failed toachieve sustained positive outcomes and, based on recent statis-tics, these policies have fallen short of improving the quality of lifefor thousands of ethnic minorities in the northern highlands. Theseoutcomes are symptomatic of planned interventions that do notaddress the agro-ecological and socio-cultural complexities of thisregion.

Officials at Hà Giang’s Department of Agriculture and RuralDevelopment (DARD) noted that hybrid maize seeds were firstintroduced to the province in 2001. Provincial and district levelgovernment officials (and documents) argue that agriculturalscience and technology, intensive cultivation, and new plant vari-eties (including hybrid maize) will improve economic develop-ment in the province (People’s Committee, Hà Giang Province,2012). However, these seeds are only available through cash trans-actions from either government distribution centers (via agricul-tural extension officers) or from suppliers in local marketplaceswho have privately imported seeds from China.

Elsewhere in these uplands, local people’s resistance5 and inno-vative livelihood coping strategies in the face of such agriculturaltechnologies are increasing. Bonnin and Turner (2012) note in theirGeoforum study of hybrid rice in neighboring Lào Cai province thatfarmers experiment with different rice seeds to maximize yieldsand taste. A selective diversification approach is the most popularagricultural livelihood strategy there: local rice varieties are pre-ferred for their taste and are grown for as long as possible, whilehybrids are turned to only when necessary for their higher produc-tivity (often due to land constraints). Yet farmers complain that

5 The forms of resistance that we are suggesting here are covert in nature, given thepolitical context in which these farmers live, and majority-minority circumstances(see Turner et al. 2015). Often undetected by officials, these resistance measuresparallel those explored by Kerkvliet (2009) as part of his ‘everyday politics’.

hybrid rice seeds are unreliable in upland agro-ecological conditions,and combined with state management of the hybrid program, farm-ers must find resourceful ways to compensate for late delivery times,uneven supplies, and poor seed choices (Bonnin and Turner, 2012).Such struggles with hybrid rice in a neighboring province, yet onewith a more forgiving topography, lead us to delve further into thehybrid maize program and local peoples’ reactions in Hà Giang.

4. Yielding to high yields?

4.1. Livelihood transforming structures: the roles of state actors in theadoption of hybrid seeds

The Vietnamese government has been instrumental in opera-tionalizing the agrarian transition, in turn impacting upland liveli-hoods through a series of institutions, organizations, policies, andother legislation; namely the ‘transforming structures and pro-cesses’ of the livelihood framework (DfID, 1999). In addition to Pro-gram 135, discussed above, which directly impacts ethnic minoritycommunities, hybrid seed management is organized nationally bythe Ministry of Agriculture and Rural Development (MARD), whichoversees a Crop Department (including a National Center for PlantVariety Testing and Certification), the Plant Protection Department,the Technology Department, and the National Center for Agricul-ture Extension. Provincial level seed management at the Depart-ment of Agriculture and Rural Development (DARD), and Districtlevel seed management mirror the national organization in crop,extension, and plant protection offices. Government agriculturalofficials collaborate with seed companies to undertake research tri-als assessing seed quality and yields.

To date, hybrid maize seeds are supplied by two main seedcompanies: Syngenta, a Swiss agrochemical and seed conglomer-ate; and Charoen Pokphand (CP) seed company, an intermediarysupplier based in Thailand (Hà Giang DARD officials, pers. comm.).6

These seed companies dictate the prices at which the governmentcan purchase hybrid seeds, and in turn the government controls seedprices for farmers. Every district in the province has a seed distribu-tion centre selling seeds. Despite state attempts to regulate prices,officials indicated that prices are higher in rural districts furtherfrom the provincial capital city.

Extension service delivery to farmers occurs through commune-level agricultural extension officers (AEOs) and district level agri-cultural officials. Considering the directional flow of seeds andinformation from private industry to the provincial and then dis-trict levels, extension officers themselves have little influence overwhich specific seeds are introduced or any effective avenue tocommunicate cultivation challenges farmers are facing back upthe chain of command. Indeed, the prevalence and consistency ofcultivation challenges Hmong farmers reported to us concerninghybrid seeds that have been cultivated for a number of years, high-lights the lack of effective feedback mechanisms. At the communelevel, agricultural extension services supposedly assist local farm-ers with cultivation concerns regarding hybrid maize seeds. Yetaccess to technical seed information rests heavily on meetings usu-ally held in the Vietnamese language, not local languages such asHmong. This limits the knowledge attendees gain, as well as whatinformation can be transferred to family members not attendingdemonstrations. Extension support also depends on the numberof households an AEO serves, and the degree of trust between AEOsand local farmers. Providing extension services to a large numberof scattered village populations is challenging, with one AEO inter-

6 The seed and agrochemical multinational company Monsanto is not yet associ-ated with hybrid maize seeds in these uplands. This does not preclude its involvementin the future however, given the company has offices in Ho Chi Minh City.


viewed in Phố Cáo commune serving 91 households, a fairly com-mon constituency size (AEO, pers. comm.).

Interviews with extension officers throughout communes inÐồng Van district revealed that households had variable rates ofhybrid maize seed adoption (corroborated by our interviews withfarmers themselves). The adoption rate was slow in Phó Bảng com-mune, as Hmong AEO Mr. Cai7 explained, with Hmong householdscontinuing to rely on local maize to supply their food and feed needs.Mr. Kau, a junior Hmong agricultural extension officer in Tả Phìncommune, admitted the most challenging aspect of his first sixmonths on the job was trying to convince Hmong farmers of theadvantages of new hybrid maize seeds: ‘‘Some people believe me,but some say ‘you just coming out of school and we have been doingthe farming for all our life’”. Conversely, Mr. Phuc, a Kinh extensionofficer, described hybrid maize adoption in Lung Cú commune as aforced process that took place over a number of years:

At first they didn’t know how to use it exactly, and from 2008we had to really press the local people to plant it. Right now,they are already used to the hybrid corn and now we don’t needto order them to plant hybrids, they actually come to the com-mune to get it themselves, but before we had to really pressurethem to plant.

Understandably, local farmers are apprehensive to change theirlong-held methods of maize cultivation, and mistrust of extensionofficers with limited practical farming experience remains high.Moreover, while coercion by agricultural extension officers reflectsthe state’s desire to impose agrarian change onto local households,Hmong farmers’ reactions, detailed next, highlight their ability tochallenge this imposition, consistent with the overall risk averse-ness of semi-subsistence farmers (cf. Just and Zilberman, 1983;Besley and Case, 1993; Feder and Umali, 1993).

Interviews with government officials highlighted that the avail-ability aspect of conceptualizing food security is often overempha-sized in agricultural policies for poverty reduction in theseuplands. Unmistakably, those working at the provincial Depart-ment of Agriculture and Rural Development equate increasingmaize availability via hybrid maize with improving livelihoods.As one official put it, ‘‘maize seed is very important – increasedquantities of maize helps people have better lives” (Hà Giang DARDOfficial, pers. comm.). While increasing food supply is undoubtedlya critical aspect of food security, agricultural policies directedsolely at increasing production can fail to sufficiently address thesocio-cultural implications of high-yielding agriculturalinterventions.

It is important to note though, that in a one-party socialist polit-ical system such as Vietnam’s, where local policies cannot contra-vene national decrees, local commune level officials ‘on theground’ do express ‘‘rightful criticism” of the implementation pro-cess of national and provincial policies they must officiallyendorse, but have no role in designing (Turner et al., 2016). Alsoworking in Hà Giang province, Turner et al. (2016) found evidencethat provincial level decrees and policies generally fail to considerthe local context in which officials operate, ignoring factors such asa diverse and difficult topography, poor infrastructure, and limitedroad access. Indeed, many of the agriculture extension officers(predominantly Hmong) we interviewed voiced critiques of hybridmaize similar to those reported by Hmong farmers including poortaste, insect infestations, and spoilage during post-harvest storage,as discussed next. These repeated concerns suggest that the gov-ernment is applying lowland surplus-driven agriculture policies,in the form of high-yielding maize seeds, to an upland environ-

7 All names are pseudonyms and are first names with an additional gender marker.All quotes are from 2013 fieldwork.

ment and cultural context most likely unsuitable for intensiveagriculture.

Moreover, NGO representatives are concerned about how iso-lated policy initiatives create an artificial separation between areasof responsibility for different departments, ultimately creating pol-icy or program silos instead of a coordinated policy framework. InVietnam, such policy failures frequently result in a disconnectbetween national government departments and their respectiveprogramming areas (World Bank, 2009a; Turner et al., 2015). OneNGO representative highlighted this clearly in a 2013 interview:

When I say MARD [Ministry of Agriculture and Rural Develop-ment] is responsible for income-generation, and MOLISA [Min-istry of Labour, Invalids, and Social Affairs] is responsible forhuman resource development, you can see the problem of pol-icy because you cannot separate farming from humanresources. For example. . . I say MARD should change the waythat it provides more capacity building for human resources,but they say ‘no no no, that’s not my business, that’s MOLISA’. . .MARD has a very small budget for training, it is very much adhoc training. . . and they cannot ask for more budget becauseMOLISA holds all the budget for human resources. . .andCEMMA [Committee for Ethnic Minorities and MountainousAreas] is responsible for culture and customs; you cannot sepa-rate customs and culture from agriculture budgets!

All told, while introducing hybrid seeds to the northern uplandsis part of a wider national state strategy to modernize and intensifyrural agricultural systems, it also appears that the official push forhigh-yielding maize adoption among Hmong households in HàGiang province is viewed as a relatively simple agronomy-basedsolution to food insecurity.

4.2. Farmer perspectives: Hybrid seed adoption and Hmong livelihoodvulnerabilities

Dilley and Boudreau (2001) argue that ‘vulnerability’ withinfood security debates is often defined in terms of specific outcomessuch as hunger, food insecurity, or famine. This can mask socio-economic or ecological events or shocks that can lead to such out-comes (but see Sen, 1981 as an exception). By contrast, vulnerabil-ity analysis within a livelihoods framework tends to focus on thespecific causal factors that can have direct impacts on people’slivelihood assets and options (DfID, 1999). We focus on four suchfactors here.

4.2.1. Natural and financial capital demandsA critical delineation has been introduced by the hybrid seed

program: only those households with access to relatively flat landcan plant hybrids, while local seeds continue to be grown in smallpockets of soil on steep karst landscapes (the soil often initiallyplaced there by farmers themselves) (Fig. 2). Provincial agriculturalofficials and farmers could not provide any insight into why hybridmaize is only viable when cultivated on flat land. Perhaps it isbecause early hybrid maize research trials were completed in low-land environments in Vietnam or elsewhere, and more recently, inaccessible, flat areas of Hà Giang.

Interviewees with access to the required natural capital, namelyflat land, noted that seed subsidies are important for obtaininghybrid maize seeds. These subsidies are a key affirmative tool inthe state’s agricultural policy toolbox. Subsidies are granted basedon household income or poverty level following Program 135, andminority farmers pay anywhere from 20% to 100% of the seed price.As Hmong farmer Ms. Hao explained:

The government supports some households 100%, especially thepoor families, while the normal [middle-income] families and

Fig. 2. Flat landscapes where hybrid maize can be grown (left), contrasted with steep kart landscapes where only traditional landraces grow (right).


the rich families, they have to pay 100%. It also depends onwhich seed you have; some seed is a little bit more expensive,some is less expensive. CP999 is the cheapest, but a lot of peopledo not have enough money to pay for this. . .

Interviewees added that both the prices of seeds and the level ofsubsidy a household is entitled to, can change from year to year,creating livelihood uncertainties and stress over financial capitalneeds.

Since hybrid maize cannot be used to propagate new seeds,farmers must bear the expense of buying new seed, subsidized ornot, every year (there is generally one maize crop annually in thesehigh-elevation communes). Hmong householders, such as Mr. Yeein Tả Phìn commune and Mr. Nhia in remote Lung Cú commune,reported frequently selling alcohol distilled from hybrid maize,along with pigs reared on hybrid maize feed to generate this nec-essary financial capital. Mr. Yee’s household plants on average1 kg of hybrid maize seed annually. Being classified as ‘‘normal”(middle income) by local officials, his household does not qualifyfor seed subsidies and must pay the full seed price. Similarly, Mr.Nhia, having a highly productive farm planting on average 20 kgof hybrid maize seed annually, pays the full price for hybrid maize,and like Mr. Yee, was also selling pigs to generate the requiredincome. A few farmers also gain employment locally or acrossthe border in China as agricultural labourers, or sell homemadetextiles in local markets, to access necessary cash. Money is alsoneeded because, despite the agro-ecological conditions making itimpossible to grow terraced wet-rice and tough to even grow dryrice, Hmong households overwhelmingly prefer to eat rice as theirstaple diet food rather than maize, buying rice from local marketswhere accessible.

When growing traditional maize landraces, local farmers mixorganic fertilizer – known locally as ‘black fertilizer’ – from ashand animal (usually cattle) dung. To grow hybrid maize, farmersare encouraged to use chemical fertilizers (known locally by thefertilizer’s color: blue, white, green, red) and apply approximately70–75 kg per application. Refusing to completely abandon tradi-tional ecological knowledge however, farmers tend to use both:‘black fertilizer’ is applied when the seeds are planted, followedby one or two chemical fertilizer applications during the plants’growth cycle. For instance, Mr. Yee, a 29 year old farmer withtwo children in Tả Phìn introduced above, applied a mixture of nat-

ural and chemical fertilizers three times over the maize growingseason to improve root development. Similarly, Mr. and Mrs. Sua,both in their early 50s with five school-aged children, cultivatedhybrid maize in Phó Bảng by applying the mixture twice season-ally. While hybrid maize seeds are subsidized by the state for somehouseholds, chemical fertilizers are not – despite being a co-requisite to hybrid seed cultivation. Moreover, fertilizer pricesare high, with farmers in Phó Bảng commune, Ðồng Van district,purchasing fertilizer in Chinese border markets for VND400,000–450,000 (USD$20–22.50) per 50 kg sack. The cumulative cost ofmultiple fertilizer applications for hybrid maize over the growingseason is a significant drawback. As agricultural extension officer,Mr. Phuc, explained: ‘‘The hybrid maize produces more maizeand of course changes the local people’s life and they harvest more.But on the down side, they use a lot of fertilizer, and they can onlygrow [hybrid seeds] on flat land, not on the hills”.

Worrying about having enough cash at the right time to accessseed and fertilizer stocks impacts material and psychological well-being, both important when conceptualizing food security. AsHmong farmer Mr. Xa, father of two young boys living in Tả Phìncommune, explained: ‘‘We have to worry about selling animalsto get money to buy the seed, otherwise if you don’t order it, andyou don’t pay for it [at the necessary time], you won’t get anything[seed] to grow”. Provincial level agriculture official Mr. Cuong atDARD added, ‘‘it is difficult for Hmong to make investments innew seeds – they are expensive and conflict with traditional farm-ing practices”.

Farmer interviewees noted that incorporating hybrid maize intotheir livelihood portfolios as an alternative to lower yielding localmaize has generally increased the quantity of maize produced(i.e. the availability aspect of conceptualizing food security).Hmong farmer Mrs. Lan explained that growing hybrid maizemeans she and her husband have enough to eat annually, thanksto effective chemical fertilizers and higher harvests. Yet, ratherthan switching completely to hybrid seeds on their suitable landin Tả Phìn, this small farming household (without children) hasbeen growing hybrid maize for the past five years following a50:50 hybrid:local proportion due to the cash demands of hybrids,their land suitability, and other concerns discussed next. Indeed,just over half (54%) of the households we interviewed that weregrowing hybrids have resisted the full adoption of hybrid maizeseeds in a selective and partial adoption strategy.

8 This mirrors recent findings in neighbouring Lào Cai province, where local Hmonghouseholds also preferred the taste of traditional Hmong rice varieties over hybrids(Bonnin and Turner 2012).


4.2.2. Hybrid variety concerns and pest pressuresOftentimes, cultivation challenges force households to switch

from one hybrid maize variety to another. For instance, Mr. andMrs. Sua in Phó Bảng commune, introduced above, cultivatedhybrid seed variety CP999 for over 15 years, but switched toCP555 due to declining root development. Likewise, farmers Mr.Xee and Mr. Txoo explained that their households switched fromgrowing variety NK4300 because the weather in their commune,Tả Phìn, is ‘‘not good for this type of seed”. Due to such problems,these households prefer to growmore local than hybrid maize, rec-ognizing that local seeds are more drought resistant.

The susceptibility of hybrid maize to spoilage during storage isalso critical to adoption rates. Farmers Mr. and Mrs. Vaa explainedthat hybrid maize ‘‘goes bad [rots] really quickly. . .so when it’sgood, it’s better than local maize but when it’s bad, its much worsethan local maize”. Householders reported various durations hybridmaize could be stored before spoiling, with these periods alwaysshorter than for local maize. For example, Hmong farmer Mr. Xa,who cultivates CP999 hybrid maize seeds, noted that hybrid maizecan be kept for up to one year, while local corn can be stored twiceas long. Mr. Tu, a Hmong police officer and farmer in Phố Cáo,praised high-yielding hybrid maize, yet admitted it decays quicklywithin three months, adding, ‘‘many people don’t want to growthis corn because it’s too difficult to keep”. In Tả Phìn, elderlyfarmer Mr. Xee and a young farmer Mr. Txoo linked their prefer-ences for local maize over hybrid maize to both taste and storageproperties, with Mr. Xee explaining: ‘‘We like the local corn more,because it tastes better and we can store it longer”. For thesehouseholds, longer storage stability of local maize was attributedto its stronger husk compared to the thinner, weaker husk ofhybrid maize. For some, the livelihood vulnerabilities of adoptinghybrid seeds are just too great, with the households of Mr. Thoand Mrs. Tsa eventually abandoning hybrid maize cultivation dueto its susceptibility to spoilage.

A Kinh agriculture official attributed such storage problems tothe so-called ‘traditional way’ of Hmong farmers and their slow-ness to adapt: ‘‘We teach the local people how to store it but theyare still living in a really traditional way. . .people are used to stor-ing it upstairs in their house, and the smoke coming from down-stairs [the kitchen fire] is sometimes not dry enough and that’swhy it goes bad really quickly”. She added that traditional storagepractices are not good for hybrid maize, saying ‘‘you have to take alot of time to break it down into small pieces, make it really dry,and put it in a big sack; that’s the proper way to store it”. Whilethis might be the case, with limited information from agriculturalextension officers, farmers need to find their own solutions andmaintaining traditional varieties is a rational approach for many.

Adding to storage concerns, post-harvest pest managementstrategies are critical to maize yields and food security, asmiddle-aged Hmong farmer Ms. Chi in Phố Cáo communeexplained: ‘‘A lot of bugs eat the corn because the leaves that growaround the corn cob are not long enough to cover the entirecob. . .and that’s why also when it rains, all the water is coming intothe corn, so you really have to harvest it at the right time, in theright season”. Likewise, Kinh agriculture extension official Mr. Phucnoted ‘‘after harvest it is difficult to store [hybrid maize] becausethe end [of the husk] is always open a little bit and then is alwaysvery easy for the insects to get in it and to eat. . .”. For this specificconcern, agricultural extension officers were able to provide littlerecourse.

4.2.3. Declining intercropping and agrobiodiversityMixed cultivation and intercropping are typical features of

Hmong food systems that increase the productive output of limitedlandholdings. Local maize plants are grown further apart thanhybrid plants, allowing space for intercropping pumpkin or beans,

contributing to greater overall farming output. Moreover, cultivat-ing beans and other legumes in mixed cultivation or crop rotationis widely understood to benefit soil fertility by encouraging theactivity of nitrogen-fixing microbes (Cassman et al. 2002, Tilmanet al. 2002, Olivares et al. 2013). However, as Hmong farmer Ms.Lang explained: ‘‘When we grow hybrid corn we cannot growbeans in between, so we only grow hybrid corn. . .because the dis-tance is too close, 20 cm is not enough to grow beans”. This limitedability to intercrop means households must access, prepare andcultivate more land, or risk reducing total farming output.

Agrobiodiversity maintained by small-scale farmers is a corner-stone of global food security, broadening the varietal database ofplants and reducing vulnerability to climate stresses and shocksand pests (Brookfield, 2001; Liang et al., 2001). Such agrobiodiver-sity is a concern in Hà Giang province, since it is near impossible tocontrol how the genetic components of hybrid plants interact withnative species planted in close proximity. As the dominance oflocal maize genotypes and phenotypes becomes somewhat diluted,it becomes difficult to distinguish between local and hybrid types.Not surprisingly, the mixing of local maize varieties and newerhybrid maize plants is evident when speaking with Hmong house-holders and maize vendors in marketplaces. One elderly farmerand military veteran in Phó Bảng, Mr. Tee, noted how cross-pollination between hybrid and local maize varieties is contribut-ing to the loss of local varieties:

Now we have the local maize but when we grow it together[with the hybrid maize] the flowers go through each other soit’s all mixed. . .and the taste is not very good for the hybridmaize. Now it’s difficult to find local maize because they areall mixed. . .because we grow too much hybrid maize, it isalready mixed all together.

As Vernooy (2003, 3) has argued, ‘‘modern agriculture is like ahuge inverted pyramid; it rests on a precariously narrow base”.This narrow base is partially comprised of a small number of vari-eties designed for intensive production. Cross-pollination and mix-ing of local and hybridized agricultural plants is a naturallyoccurring phenomenon with implications for seed diversity, andby extension, food security. Cultivating monocultures of foodplants, whether through selective breeding (hybrid seeds) orbiotechnology, is a controversial strategy in modern agricultureoften associated with losses in agrobiodiversity over time, suscep-tibility to disease, and soil infertility (Altieri and Merrick, 1987;Altieri, 1999; Zhu et al., 2000; UNEP, 2005; Smith et al., 2008).Indeed, Shiva et al. (2002) argue that rejuvenating agrobiodiversityis the most sustainable insurance against pest damage. Ultimately,the Hà Giang provincial government’s strategy to replace localvarieties of maize with high-yielding hybrid maize will have farreaching impacts on the diversity of local upland food systems.

4.2.4. Taste preferences and food acceptabilityTalking to farmers it became increasingly clear that conceptual-

izing vulnerability to food insecurity in terms of outcomes such asextreme hunger was overlooking a crucial conceptual element offood security and livelihood equations in the region, namely cultur-ally acceptable food. Taste preferences directly influenced howhouseholds decided to use hybrid maize, and in addition to prefer-ring rice to maize, Hmong householders voiced clear taste prefer-ences for their traditional maize over hybrid varieties.8 Hmongfarmer Mrs. Tsa explained, ‘‘the taste of [hybrid] maize is not reallyas good as regular local maize that we have. Hybrid maize is only forfeeding pigs, and especially for making alcohol. . . So we don’t eat it


so much. . .we eat the local maize more”. She clarified that her familyretains access to local maize as a staple, despite its lower yields. Sim-ilarly, farmer Mr. Xa noted that his family uses hybrid maize ‘‘forlivestock and also to eat. . .but we will eat the local maize more thanthe hybrid corn. . .the local corn tastes much better”.

Farmers thus divert the less palatable hybrid maize to livestockfeed and alcohol production as a strategy to retain more of thesuperior tasting local varieties for household consumption. AHmong woman selling maize kernels at Ðồng Van market,explained that local maize has a softer texture, and therefore‘‘tastes much better” than the tougher hybrid maize. These differ-ent properties of traditional versus hybrid maize surfaced againand again during interviews, suggesting a strong associationbetween taste and food acceptability. Indeed, Hmong farmers’comments on taste were unanimous, reflecting a highly embeddedcultural response and perhaps even a cultural marker of distinctionfrom lowland populations. Further, farmers frequently spoke posi-tively of those they knew who had enough land to be able to main-tain traditional varieties of maize.

5. Discussion: The roles of agency and cultural preferences

Hmong farmers in these uplands are exercising their agency tomoderate the impacts of hybrid maize policies on their livelihoods.Mrs. Ze, a successful maize alcohol vendor in Ðồng Van district,highlighted this agency when sharing her perspective on the stateseed subsidy strategy: ‘‘The government says we are very poor, andthey want us to leave the local corn, not to grow the local corn any-more, so that’s why they give us the free [hybrid] seed to grow forthe land here, but we only grow it for pigs and for alcohol, not toeat”. Likewise Hmong farmers retain local maize varieties wher-ever possible to reduce household vulnerability to yield lossesfrom pests known to attack genetically similar monocultures(Altieri and Merrick, 1987; Altieri, 1999). Such subtle resistanceis confronting planned interventions that have been based on anoversimplified linear model of policy, implementation, and posi-tive agrarian outcomes (Long and Van Der Ploeg, 1989). The resultis ‘‘the reinterpretation or transformation of policy during theimplementation process [by Hmong farmers], such that there isin fact no straight line from policy to outcomes” (Long and VanDer Ploeg, 1989, 227). Households organize themselves in differentways in the face of state-planned interventions – at the ‘socialinterface’ between structures and actors (Long, 2001) – and whennecessary subtly challenge or quietly resist the full adoption of thisnew agricultural technology.

The experiences of upland Hmong farmers in Ðồng Van areechoed by upland maize farmers in neighboring China, where Liet al. (2012) have highlighted the persistence of landraces plantedfrom farmer-saved seeds on smallholder farms in the karst moun-tainous region of southwest China (Guangxi, Yunnan, and Guizhouprovinces). This trend has occurred despite the rapid expansion ofhybrid maize cultivars, new seed regulations (in part to promotehybrid seed adoption), and the establishment of a commercial seedmarket. Likewise, in Guatemala’s western highlands, van Ettenet al. (2008) found maize diversity persists (despite the influenceof modern varieties) due to local and regional seed exchanges thatare a source of crop innovation. Moreover, Negi (1994) found in themountainous Himalaya region of India that farmers did not adoptwheat high-yielding varieties (HYV) despite marginally higheryields, because corresponding declines in straw yields did not meettheir livestock feed needs.

In Vietnam, theWorld Bank (2009a, 243) states that inappropri-ate development policies designed by lowland Kinh continue tomisinterpret minority livelihoods and cultures, suggesting thatminorities are intellectually inferior and must be shown ‘‘how to

develop”. McElwee (2004) adds that ethnic minorities have beenquietly resisting poorly planned development policies formulatedat the centralized national state level, such as monocropping meth-ods encouraged in highland areas. Further, Turner et al. (2015) pro-vide evidence of a range of ways Hmong farmers indigenizemodernity and quietly resist the wholesale acceptance of thestate’s ‘development’ approach.

As these cases briefly show, care must be taken in livelihoodresearch not to ‘‘assume that people are entirely dedicated to max-imizing their income. Rather, we should recognize and seek tounderstand the richness of potential livelihood goals. . .to under-stand people’s priorities, why they do what they do, and wherethe major constraints lie” (DfID, 1999). Hence, what do Hmonghouseholder experiences in Hà Giang tell us about people’s liveli-hood priorities and how this agrarian intensification program hasimpacted food security outcomes?

The benefits of planting hybrid maize, namely higher yields forhuman food and a feed source to support more livestock (thereforeincreasing potential wealth accumulation), are tempered in HàGiang province by a range of challenges and vulnerabilities. Theseinclude inconsistent seed subsidies, the yearly cost of buying seedsand fertilizers, the specific type of land required, the inability tointercrop hybrid maize with other food staples, and the suscepti-bility of hybrid maize to spoilage during post-harvest storage bymold or insect infestations. Hybrid maize adoption is also associ-ated with an unexpected outcome relating to food acceptability,namely hybrid maize’s inferior taste. As such, hybrid maize alcoholand livestock reared on hybrid maize-based feed are sold amonghouseholds or at local marketplaces for cash income, used in turnto purchase hybrid maize seed inputs and rice to eat. This indirectuse of hybrid maize to purchase rice was a surprising, yet recurringtheme in three-quarters of our household interviews, and furtherillustrates the significance of food preferences in influencing liveli-hood strategies and how livelihood assets (especially financial cap-ital) are used.

Some twenty percent of householders interviewed reportedhaving ‘‘a better life” with hybrid maize yields, an intangible out-come of well-being, yet an equal proportion of adopters perceivedthemselves as remaining food insecure, with no increased benefits,and still not having enough food to eat throughout the year. Thus,despite the introduction of hybrid maize, food insecurity persistsamong some Hmong households in Ðồng Van district. Household-ers’ experiences with new varieties also suggest the possibility thathybrid maize quality and productivity decline over time, causingdependence on a continuous supply of newer ‘improved’ seeds.An agro-technology adoption treadmill may result, where depen-dency on new varieties compounds the permanent losses of localindigenous maize agrodiversity – a looming threat to future foodsecurity for Hmong farming households.

In ignoring farmer concerns over subsidies, input costs, maizestorage, pests, preferred taste, and agrodiversity, the central andprovincial Vietnamese governments fail to recognize what Longand Van Der Ploeg (1989, 228) refer to as the fundamental natureof intervention; ‘‘an ongoing, socially-constructed and negotiatedprocess, not simply the execution of an already-specified plan ofaction with expected outcomes”. From the perspectives of farm-ers, there are important limitations to theoretical models ofagrarian change that drive policy intervention practices. One suchlimitation is that intervention ‘‘is not a discrete phenomenon inspace and time”, but rather is linked to the ‘‘historical imprint”of past interventions and local context (Long and Van DerPloeg, 1989, 229–230). With these aspects in mind, revisitingthe sustainable livelihoods framework highlights how state-supported hybrid maize seeds are testing the flexibility of uplandHmong livelihoods, and at times increasing their vulnerability.Diagrammatically, this can be seen by superimposing the realities

F

S

H

N

P

Influence&Access

LIVELIHOODOUTCOMES

Food security – more maize,

access to rice

Food insecurity – not enough

food, unreliable supply (due to

pests, inappropriate seeds)

Resistance to hybrids - 50:50

HM:LM

Discontent with HM taste

LIVELIHOODSTRATEGIES

Selling HM alcohol

Selling livestock

Micro-enterprise

Employment

TRANSFORMINGSTRUCTURES +PROCESSESCulture

Agribusiness

Government

Agriculture Extension

NGOs

Capital Assets

H = Human S = Social

N = Natural P = Physical

F = Financial

LIVELIHOODASSETSFinancial – cash for seeds + inputs

Natural – flat land for HM

Physical – livestock for cash

VULNERABILITYCost of seeds +fertilizer

Sustainability of HM seed subsidies

Reliance on cash + livestock bank

Cannot intercrop with HM reduced

farm output

Need flat land for HM

Post-harvest spoilage

Loss of local maize diversity +

traditional seed saving knowledge

Fig. 3. The Livelihoods Framework for Hmong Households adopting hybrid maize in Ðồng Van District, Hà Giang (adapted from DfID, 1999). HM = hybrid maize, LM = localmaize, 50:50 refers to the predominant maize cultivation strategy.


of the hybrid maize experience for Hmong households in ÐồngVan district on the livelihoods framework (Fig. 3).

Hmong households adopting hybrid maize experience alivelihood vulnerability context associated with an increasedneed for financial capital (reliance on cash to purchase seeds,fertilizers and pesticides), limitations to traditional farmingtechniques (a decline in intercropping), and the need for flatland (natural capital) to cultivate hybrid maize. Seed affordabil-ity and potential losses in local seed diversity as adoption con-tinues, are also important factors. A greater maize supply allowsmany families to generate income to buy rice – an indirect pos-itive impact on food security. However food insecurity persistsfor other households as hybrid maize does not ensure enoughfood for the year. Strategic, cautious resistance to full adoptionof hybrid maize, reflected in common 50:50 proportions oflocal:hybrid cultivation, along with the use of hybrid maizewhen necessary to generate income to buy rice, strongly suggestthe on-going cultural value of and preference for local varieties.In turn, these approaches reflect a reluctance among farmers tofully accept the state’s agricultural development policies or fulldependency on the market economy – outcomes unanticipatedby the state.

We would further argue that, as a conceptual tool, the sustain-able livelihoods framework would be enhanced by emphasizingculture as a significant transforming process shaping livelihooddecisions (including agronomical decisions) of rural subsistencefarmers. Some authors have suggested that ‘cultural capital’ shouldbe added as an additional asset to the sustainable livelihoodsframework (rather than sometimes subsumed under social capi-tal). They draw on the work of Pierre Bourdieu who notes that cul-tural capital (simply put), relates to the skills, knowledge,education, and possible social advantages a person has that cangive them advantages in society (Bourdieu, 1984). Yet, what weare proposing here is slightly different. We are focusing insteadon long-term culturally rooted traditions and understandings ofagro-ecological norms (such as organic fertilizer) and food prefer-ences (taste), in addition to other regular cultural constructs suchas language, social habits, religious practices, morals, and so on.Highlighting these elements is important due to the core role thatthe cultural appropriateness of specific foods plays in the story of(partial) hybrid adoption.

6. Concluding thoughts: Championing a hybrid intervention

Life in the mountainous regions of northern Vietnam is a deli-cate negotiation over state ideals of ‘market socialism’ and ‘mod-ernization’, state and lowlander prejudices of upland ethnicminorities, and the physical realities of an unforgiving, upland ter-rain. As the agrarian transition advances in this frontier region, thestate appears to be ignoring a number of long-term impacts ofintensified modern agriculture on upland livelihoods, local ecosys-tem health, and the complicated spatial and temporal dimensionsof agrodiversity. Yet the question remains: Is yielding to highyields the answer to food insecurity among highland ethnicminorities in northern Vietnam?

The World Bank (2009a, 257) warns that overemphasizingmonocropping, high inputs of fertilizer and pesticides, and hybridseeds will not serve as a sustainable model for agricultural produc-tion in ‘‘cash-poor areas too remote from markets”. The Interna-tional Assessment of Agricultural Knowledge, Science andTechnology for Development (McIntyre et al., 2009) further cau-tions that scientific achievements in agricultural knowledge sys-tems have unintended consequences that have yet to besufficiently addressed – namely persistent rural poverty due toasymmetric agricultural development between regions and coun-tries, and unsustainable use of natural resources. Offering an alter-native approach to the agrarian transition, Brookfield (2001)replaces Boserup’s linear model of intensification with an emphasison diversified production and livelihood opportunities. Likewise,within their critical appraisal of development challenges facingethnic minorities in Vietnam, the Ethnic Minority Working Group(2014) acknowledges that many minority livelihoods in theuplands will continue to depend on subsistence agriculture in thenear future, therefore recommending that policies recognize andsupport the small-scale diversified agricultural model these com-munities have long had in place. As Long and Van Der Ploeg(1989) suggest, planned interventions – in this case, in the formof hybrid seeds, chemical fertilizers, and other associated technolo-gies – are not necessarily a precursor to agrarian transformations,but rather often part of the problem of ‘development’.

To improve food security in Vietnam’s northern uplands it isnecessary to maximize productivity on limited arable land area.Yet, judging from the observed challenges to hybrid maize


adoption, alongside uplanders’ intimate knowledge of the sustain-ability of already fragile upland food systems, state efforts to solely‘maximize productivity’ appear to be stalling. Completely replacinglocal maize varieties with hybrid maize, which provincial authori-ties aim to do, is likely to make fragile Hmong subsistence foodsystems more unsustainable in the longer term, adding consider-able stress for these farmers. Clearly, this planned interventionfor agricultural development and food security does not considerthe practical, socio-cultural aspects of Hmong semi-subsistenceagriculture. Our findings hence raise important questions aboutthe future long-term impacts and sustainability of hybrid maizein these uplands.

We therefore argue that a hybrid solution is necessary – an inte-grated complementary system of hybrids and landraces supportingboth commercial and subsistence needs. As Bellon and Hellin(2011, 1442) note: ‘‘The challenge is to develop agricultural mod-ernization pathways that build on the complementary functionsof hybrids and landraces to improve farmers’ livelihoods, ratherthan focusing on substituting one by the other”. We suggest threecore features for this hybrid solution: first, the integration of agri-cultural research in northern Vietnam that builds on hybrid maizeseed adoption experiences in highland environments elsewhere inthe rural Global South – experiences that are relevant to semi-subsistence livelihoods. Second, participatory plant breeding(PPB) that engages (instead of marginalizing) local people in find-ing creative solutions for food insecurity. This approach couldmaintain, widen, and improve the seed resource database withrespect to environmental suitability, taste preferences, and con-serving seed agrobiodiversity for food security (Vernooy, 2003;Mba et al., 2012; Winarto, 2011). Indeed, the potential of thisapproach is already being explored across the border in China, asdescribed by Li et al. (2012) in their conceptual model for partici-patory hybrid maize breeding procedures. Involving farmers inhybrid development early in the pre-breeding stage broadens thepopulation base of farmer-maintained local landraces, and cansupport the co-evolution of on-farm genetic resources with hybridvarieties. Finally, food security policies must move beyond concep-tualizing food security as a result of food availability alone, andincorporate cultural acceptability to a far greater degree. Under-standing upland ethnic minority cultures, dietary preferences,and agricultural systems (including local informal seed-saving sys-tems), is essential to improving agricultural outcomes over thelong term. Whether the Vietnamese government can adopt suchfeatures will determine if hybrid maize seeds actually lead to realand sustainable improvements in food security for upland ethnicminority households into the future, or whether a rigid, top-down approach that harms local livelihoods will prevail.

Acknowledgements

Wewould like to thank Professor PhạmVan Cự, Ms. Ái Tong, andMs. Ðinh Thị Diệu for facilitating fieldwork access and for collabo-rations in the field, and to Ms. Ðinh for illustrating the maps.Thanks to Pang for her enthusiastic research assistant work. Wealso wish to thank all the upland farmers and officials with whomwe talked. Funding from the Social Sciences and HumanitiesResearch Council of Canada is gratefully acknowledged, as arethe constructive comments from two reviewers.

References

Altieri, M.A., 1999. The ecological role of biodiversity in agroecosystems. Agric.Ecosyst. Environ. 74 (1–3), 19–31.

Altieri, M.A., Merrick, L., 1987. In situ conservation of crop genetic resourcesthrough maintenance of traditional farming systems. Econ. Bot. 41 (1),86–96.

Bellon, M.R., Hellin, J., 2011. Planting hybrids, keeping landraces: Agriculturalmodernization and tradition among small-scale maize farmers in Chiapas,Mexico. World Develop. 39 (8), 1434–1443.

Besley, T., Case, A., 1993. Modeling technology adoption in developing countries.Am. Econ. Rev. 83 (2), 396–402.

Bonnin, C., Turner, S., 2012. At what price rice? Food security, livelihoodvulnerability, and state interventions in upland northern Vietnam. Geoforum43 (1), 95–105.

Borras Jr., S.M., Hall, R., Scoones, I., White, B., Wolford, W., 2011. Towards a betterunderstanding of global land grabbing: an editorial introduction. J. PeasantStud. 38 (2), 209–216.

Bourdieu, P., 1984. Espace social et genèse des ‘classes’. Actes Rech. Sci. Soc. 52–53,3–14.

Brookfield, H., 2001. Intensification, and alternative approaches to agriculturalchange. Asia Pac. Viewpoint 42, 181–192.

Carr, E., 2013. Livelihoods as intimate government: Reframing the logic oflivelihoods for development. Third World Quart. 34 (1), 77–108.

Cassman, K.G., Dobermann, A., Walters, D.T., 2002. Agroecosystems, nitrogen-useefficiency, and nitrogen management. AMBIO J. Human Environ. 31 (2), 132–140.

Chambers, R., Conway, G., 1992. Sustainable rural livelihoods: practical concepts forthe 21st century. In: IDS Discussion Paper 296. Institute of Development Studies(IDS), UK.

Caouette, D., Turner, S. (Eds.), 2009. Agrarian Angst and Rural Resistance inContemporary Southeast Asia. Routledge, London.

Dang Thanh Ha, Tran Dinh Thao, Nguyen Tri Khiem, Mai Xuan Trieu, Gerpacio, R.V.,Pingali, P.L., 2004. Maize in Vietnam: Production Systems, Constraints, andResearch Priorities, CIMMYT, Mexico.

de Haan, L., Zoomers, A., 2003. Development geography at the crossroads oflivelihood and globalisation. J. Econ. Soc. Geogr. 94 (3), 350–362.

de Haan, L., Zoomers, A., 2005. Exploring the frontier of livelihoods research.Develop. Change 36 (1), 27–47.

Department for International Development (DfID), 1999. Sustainable LivelihoodsGuidance Sheets: Section 2.1-2.6 Sustainable Livelihood Framework. <http://www.eldis.org/vfile/upload/1/document/0901/section2.pdf> (cited July 26 2015).

Dilley, M., Boudreau, T.E., 2001. Coming to terms with vulnerability: a critique ofthe food security definition. Food Policy 26, 229–247.

Eakin, H., Tucker, C., Castellanos, E., 2006. Responding to the coffee crisis: a pilotstudy of farmers’ adaptations in Mexico, Guatemala and Honduras. Geogr. J. 172(2), 156–171.

Erenstein, O., 2010. The evolving maize sector in Asia: challenges and opportunities.J. New Seeds 11 (1), 1–15.

Ethnic Minorities Working Group (EMWG), 2014. Critical issues in achievingsustainable development of ethnic minorities in Vietnam. In: InternationalConference on Sustainable Development and Ethnic Minority Poverty Reductionin Mountainous Regions. Thai Nguyen University, Thai Nguyen City, Vietnam.

Feder, G.U., Umali, D.L., 1993. The adoption of agricultural innovations: a review.Technol. Forecast. Soc. Change 43 (3), 215–239.

Food and Agriculture Organization of the United Nations (FAO), 2004. FoodInsecurity and Vulnerability in Viet Nam: Profiles of Four Vulnerable Groups,ESA Working Paper No. 04-11, Agricultural and Development EconomicsDivision, FAO.

Food and Agriculture Organization of the United Nations (FAO), 2015. World FoodSummit 13–17 November 1996 Rome Italy: Rome Declaration on World FoodSecurity. FAO 1996. <http://www.fao.org/docrep/003/w3613e/w3613e00.htm>(cited 2015).

Food and Agriculture Organization of the United Nations Statistics Division(FAOSTAT), 2015. Production Quantities by Country. FAOSTAT 2015.<http://faostat3.fao.org/home/E> (cited March 28 2015).

Forsyth, T., Michaud, J., 2011. Rethinking the relationships between livelihoods andethnicity in highland China, Vietnam, and Laos. In: Michaud, J., Forsyth, T. (Eds.),Moving Mountains: Ethnicity and livelihoods in highland China, Vietnam andLaos. UBC Press, Vancouver and Toronto, pp. 215–227.

General Statistics Office of Vietnam (GSO), 2010a. Result of the Vietnam HouseholdLiving Standards Survey 2010, ed. Central Population and Housing CensusSteering Committee. Hanoi, Vietnam.

General Statistics Office of Vietnam (GSO), 2010b. The 2009 Vietnam Populationand Housing Census: Completed Results. Hanoi, Vietnam: Central Populationand Housing Census Steering Committee.

Gerpacio, R.V. (Ed.), 2001. Impact of Public- and Private-Sector Maize BreedingResearch in Asia, 1966–1997/98. International Maize and Wheat ImprovementCenter (CIMMYT), Mexico, D.F..

Gerpacio, R.V., Pingali, P.L., 2007. Tropical and Subtropical Maize in Asia: ProductionSystems, Constraints, and Research Priorities. International Maize and WheatImprovement Center (CIMMYT).

Ha Viet Quan, 2009. Programme 135 – Sharing Lessons on Poverty Reduction andDevelopment Schemes for Ethnic Minorities in Vietnam, 7: United Nations.

Hall, D., Hirsch, P., Murray Li, T., 2013. Powers of Exclusion: Land dilemmas inSoutheast Asia. National University Press, Singapore.

International Rice Research Institute (IRRI), 2015. Breeding program management.Rice Knowledge Bank 2006. <http://www.knowledgebank.irri.org/ricebreedingcourse/Hybrid_Rice_Breeding_&_Seed_Production.htm> (cited July12 2015).

Jacobson, M., 2007. Food matters. J. Commun. Pract. 15 (3), 37–55.Just, R.E., Zilberman, D., 1983. Stochastic structure, farm size and technology

adoption in developing agriculture. Oxford Econ. Pap. 35 (2), 307–328.


Kanji, N., MacGregor, J., Tacoli, C., 2005. Understanding Market-based Livelihoods ina Globalising World: Combining Approaches and Methods InternationalInstitute for Environment and Development (IIED).

Kerkvliet, B.J.T., 2009. Everyday politics in peasant societies (and ours). J. PeasantStud. 36 (1), 227–243.

Kontinen, T. (Ed.), 2004. Development Intervention: Actor and Activity Perspectives.Helsinki, Finland: Center for Activity Theory and Development Work Researchand Institute for Development Studies, University of Helsinki.

Li, J., Lammerts van Bueren, E.T., Jiggins, J., Leeuwis, C., 2012. Farmers adoption ofmaize (Zea mays L.) hybrids and the persistence of landraces in SouthwestChina: implications for policy and breeding. Genet. Resour. Crop Evolut. Int. J. 59(6), 1147–1160.

Liang, L., Stocking, M., Brookfield, H., Jansky, L., 2001. Biodiversity conservationthrough agrodiversity. Glob. Environ. Change 11 (1), 97–101.

Long, N., 2001. Encounters at the interface: social and cultural discontinuities indevelopment and change. In: Long, N. (Ed.), Development Sociology: ActorPerspectives, London. Routledge, New York, pp. 73–92.

Long, N., 2004. Actors, interfaces, and development intervention: Meanings,purposes and powers. In: Kontinen, T. (Ed.), Development Intervention: Actorand Activity Perspectives. University of Helsinki, Helsinki, Finland, pp. 14–36.

Long, N., Van Der Ploeg, J.D., 1989. Demythologizing planned intervention: an actorperspective. Sociol. Ruralis 29 (3–4), 226–249.

Maslow, A.H., 1954. Motivation and Personality. Harper, New York.Maxwell, S., 1996. Food security: a post-modern perspective. Food Policy 21 (2),

155–170.Mba, C., Guimaraes, E., Ghosh, K., 2012. Re-orienting crop improvement for

the changing climatic conditions of the 21st century. Agric. Food Secur. 1 (1),7.

McElwee, P., 2004. Becoming socialist or becoming Kinh? Government policies forethnic minorities in the Socialist Republic of Viet Nam. In: Duncan, C.R. (Ed.),Civilizing the Margins: Southeast Asian Government Policies for theDevelopment of Minorities. Cornell University Press, Ithaca, pp. 182–213.

McGregor, A., 2009. New possibilities? Shifts in post-development theory andpractice. Geogr. Compass 3 (5), 1688–1702.

McIntyre, B.D., H.R. Herren, Wakhungu, J., Watson, R.T., 2009. InternationalAssessment of Agricultural Knowledge, Science and Technology forDevelopment (IAASTD): Global Report. Washington, D.C.: United NationsEnvironment Programme (UNEP).

McSweeney, K., 2004. The dugout canoe trade in central America’s mosquitia:approaching rural livelihoods through systems of exchange. Ann. Assoc. Am.Geogr. 94 (3), 638–661.

Michaud, J., 2009. Handling Mountain Minorities in China, Vietnam and Laos: fromhistory to current concerns. Asian Ethnicity 10 (1), 25–49.

Minot N., Epprecht M., Tran Thi Tram Anh, Le Quang Trung, 2006. IncomeDiversification and Poverty in the Northern Uplands of Vietnam, InternationalFood Policy Research Institute (IFPRI), Washington D.C.

Misselhorn, A., Aggarwal, P., Ericksen, P., Gregory, P., Horn-Phathanothai, L., Ingram,J., Wiebe, K., 2012. A vision for attaining food security. Environ. Sustain. 4, 7–17.

Moore, D.S., 1998. Subaltern struggles and the politics of place: remappingresistance in Zimbabwe’s eastern highlands. Cultural Anthropol. 13, 1–38.

Negi, G.C.S., 1994. High yielding vs. traditional crop varieties: a socio-agronomicstudy in a Himalayan village in India. Mountain Res. Dev. 14 (3), 251–254.

Nevins, J., Peluso, N. (Eds.), 2008. Taking Southeast Asia to Market: Commodities,Nature, and People in the Neoliberal Age. Cornell University Press, Ithaca.

Nguyen Viet Cuong, Phung Duc Tung, Westbrook, D., 2015. Do the poorest ethnicminorities benefit from a large-scale poverty reduction program? Evidencefrom Vietnam. Quart. Rev. Econ. Finance 56, 3–14.

Olivares, J., Bedmar, E.J., Sanjuan, J., 2013. Biological nitrogen fixation in the contextof global change. MPMI Molecular Plant-Microbe Interact. 26 (5), 486–494.

Pandey, S., Khiem, N.T., Waibel, H., Thien, T.C., 2006. Upland Rice, Household FoodSecurity, and Commercialization of Upland Agriculture in Vietnam.International Rice Research Institute (IRRI).

People’s Committee, Hà Giang Province, 2012. Quyết dịnh số: 2197/2012/QÐ-UBNDvề Ban hành quy chế phối hợp trong quản ly nhà nước về dang ky giao dịch bảodảm bằng quyền sử dụng dất, tài sản gắn liền với dất trên dịa bàn tỉnh Hà Giang.(Decision #2197/2012/QD-UBND to Promulgate Regulations in Co-ordination ofState Management of Registration of Transactions Secured by Land Use Rightsand Assets Associated with Land in the Province of Hà Giang.).

Pinstrup-Andersen, P., 2009. Food security: definition and measurement. FoodSecur. 1 (1), 5–7.

Polanyi, K., 1944. The Great Transformation. Farrar and Rinehart Inc., New York,Toronto.

Scoones, I., 1998. Sustainable rural livelihoods: a framework for analysis. In: IDSWorking Paper 72. Institute of Development Studies (IDS), Brighton, UK.

Scoones, I., 2009. Livelihoods perspectives and rural development. J. Peasant Stud.36 (1), 171–196.

Sen, A., 1981. Poverty and Famines: An Essay on Entitlement and Deprivation.Clarendon Press, Oxford.

Shiva, V., Jafri, A.H., Emani, A., Pande, M., 2002. Seeds of suicide: the ecological andhuman costs of globalisation of agriculture. In: Shiva, V., Bedi, G. (Eds.),Sustainable Agriculture and Food Security: The Impact of Globalization. SagePublications, New Delhi, India, pp. 169–184.

Smith, R., Gross, K., Robertson, G., 2008. Effects of crop diversity on agroecosystemfunction: crop yield response. Ecosystems 11 (3), 355–366.

Stromberg, P.M., Pascual, U., Bellon, M.R., 2010. Seed systems and farmers’ seedchoices: the case of maize in the Peruvian Amazon. Human Ecol. 38 (4), 539–553.

Tilman, D., Cassman, K.G., Matson, P.A., Naylor, R., Polasky, S., 2002. Agriculturalsustainability and intensive production practices. Nature 418 (6898), 671–677.

Turner, S., Bonnin, C., Michaud, J., 2015. Frontier Livelihoods. Hmong in the Sino-Vietnamese Borderlands. University of Washington Press, Seattle.

Turner, S., Kettig, T., Ðinh Thị Diệu, Phạm Van Cự, 2016. State livelihood planningand legibility in Vietnam’s northern borderlands: the ‘‘rightful criticisms” oflocal officials. J. Contemp. Asia. http://dx.doi.org/10.1080/00472336.2015.1028964; 1–29.

United Nations, 1975. Report of the World Food Conference – Rome, 5–16November 1974. United Nations, New York.

United Nations Environment Program (UNEP), 2005. Integrated Assessment of theImpact of Trade Liberalization on the Rice Sector UNEP Country Projects RoundIII: A Synthesis Report: UNEP.

van Etten, J., Fuentes López, M.R., Molina Monterroso, L.G., Ponciano Samayoa, K.M.,2008. Genetic diversity of maize (Zea mays L. ssp. mays) in communities of thewestern highlands of Guatemala: geographical patterns and processes. Genet.Resour. Crop Evol. 55 (2), 303–317.

Vernooy, R., 2003. Seeds That Give: Participatory Plant Breeding. InternationalDevelopment Research Centre (IDRC), Ottawa, Canada.

Winarto, Y., 2011. Weaving the diverse ’seeds’ of knowledge. Asia Pac. J. Anthropol.12 (3), 274–287.

World Bank, 2009a. Country Social Analysis: Ethnicity and Development in Vietnam– Main Report, vol. 2. In: Country Social Analysis (CSA). Washington, D.C.

World Bank, 2009b. Vietnam – First Program 135 Second Phase Support Operationfor Communes Facing Extreme Hardship in Ethnic Minority and MountainousAreas. Washington, D.C.

World Bank, 2012. Well Begun, Not Yet Done: Vietnam’s Remarkable Progress onPoverty Reduction and the Emerging Challenges. Washington, D.C.

World Bank, 2015. Vietnam. Overview. <http://www.worldbank.org/en/country/vietnam/overview> (cited June 2 2015).

World Health Organization, 2015. Food Security 2015 [cited July 27 2015]. <http://www.who.int/trade/glossary/story028/en/>.

Zeven, A.C., 1998. Landraces: a review of definitions and classifications. Euphytica104 (2), 127–139.

Zhu, Y., Chen, H., Fan, J., Wang, Y., Li, Y., Chen, J., Fan, J., Yang, S., Hu, L., Leung, H.,Mew, T.W., Teng, P.S., Wang, Z., Mundt, C.C., 2000. Genetic diversity and diseasecontrol in rice. Nature 406 (6797), 718–722.

yielding to high yields? critiquing food security...

Documents